LAUSR

Studies have found math anxiety and achievement to be related from the beginning of formal schooling, but the knowledge regarding the direction of the relationship is vague. The purpose of the present study was to study this relationship. We investigated math anxiety from two points of view: trait and state anxiety. In the first substudy, we investigated the longitudinal relationship between math anxiety and performance from sixth to seventh grade (n = 848) with cross‐lagged modeling. In the second substudy, we investigated the situational relationship of anxiety and performance by giving the participants (n = 149) challenging and nonchallenging math tasks adapted to their skill level, and then examining the association between anxiety and the performance. The results suggest that math anxiety has a small longitudinal effect on performance: High anxiety in sixth grade predicted low performance in seventh grade. Anxiety also had a situational association with performance: when anxiety was aroused, the participants performed more poorly compared to their skill level. The results adduce the two‐fold effect of anxiety on achievement: math anxiety seems to have both a real‐time association with performance and a long‐term effect on the development of basic arithmetic skills.